First-Hand:Generators and Electrical Insulation

Submitted by Vernon McFarlin

June 15, 1931-graduation into twenty five percent unemployment. If one had a college degree and a dime, one could ride on the Boston Elevated Railway.

I had worked part-time jobs and borrowed money to train myself for skilled employment. After four years of struggle, graduating into a world where jobs were practically non-existent was an embittering experience. I will never know what it feels like to be offered even one job upon graduation from college. Although I was unemployed for a total of only about five weeks during the Great Depression, the work I had paid only subsistence wages.

At the AIEE Winter Power meeting in 1959, I presented Conference Paper CP58-1310, "Tests and Life Expectancy of Generator Windings." This paper was listed among other references in a paper by a British engineer. I received a copy of the paper from an associate who maintained hydro generators for the Aluminum Company of Canada. It described the only method I had ever heard of for using test results to predict the life of electrical insulation. Some tests the author used were more sophisticated than Boston Edison's, but I still used his system of evaluation on the tests we routinely made on our generators.

On Friday, September 20, 1963 (the day I was fifty-four years old), I inspected one of the Edison Company's large generators and recommended removal of the field so I could make a more thorough inspection. Although this would involve spending probably well over ten thousand dollars, my recommendation was accepted without question.

After the field was removed, I was able to inspect the armature winding. It showed some signs of deterioration. The manufacturer's engineer inspected it and said to me, "Mac, that generator is thirty-five years old and looks it. It should be rewound."

I said, "We'll give it a high potential test and if it passes, we'll put it back to work."

The engineer responded, "You're taking an awful chance." It stood the test. The microampere leakage readings indicated a winding in better than normal condition for a machine that old and I told my bosses to put it back together. A rewind would have cost over one hundred thousand dollars.

At the Electrical Insulation Conference in New York in September, 1965, I presented IEEE paper 32C3-90 entitled, "Evaluating the Condition of Generator Stator Insulation by Test." I presented an updated report based on more data to the Doble Client Conference in 1973. An engineer from Ontario Hydro telephoned me after the Doble Conference presentation. He said it was the only reference he had found which showed how test results could be used to predict the life of electrical insulation. He was interested in how it might be applied to his company's generators.